Mass spectrometry

During the last ten years, proteomics has switched the analytical platform from gel-based to mass spectrometry-gel-based techniques [63, 89, 99]. In the years just after 2000, optimism was great to “high-throughput methods” such as MALDI-TOF, short for matrix assisted laser desorption/ionisation – time of flight mass spectrometry, and SELDI, which is a version of the same with specific surfaces to improve ionization.

MALDI is a method to make ions enter a mass spectrometer, introduced by Karas et al [101, 102]. Briefly, the sample is placed on a surface and mixed with a matrix for ionization, often consisting of small organic acids such as sinapinic acid or other substituent molecules from cinnamic acid. While the precise mechanism is not fully understood [103], the principle is that the dried sample-matrix spot is ionised with a laser, resulting in vaporised charged particles able to enter the mass spectrometer.

MALDI is a stable, quick and simple method for analyzing samples. As a method, it does not permit chromatography in itself; this must be performed off-line before spotting. In general the mass resolution, meaning the instrument specific ability to separate two specific similar masses, of a MALDI-TOF is not high enough for high confidence peptide identification [103].

Several research groups developed over many years MALDI as a principle and method for ionization. Koichi Tanaka was awarded a part of the 2002 Nobel Prize in chemistry for demonstrating that a combination of laser wave length and energy to a matrix with corresponding physical and chemical properties could cause a soft ionization [104], a prize John Fenn also took a part in for the development of electrospray ionization (ESI).

Electrospray ionization mass spectrometry (ESI-MS), introduced by Fenn et al in 1989 [105], has enabled analyzes of samples in a mass spectrometer without being

dependent on matrixes or specific surfaces for ionization. The principle is as for MALDI to perform a soft ionization of peptides, where soft meaning ionization without breaking structural chemical bonds – as opposed to hard ionization with fragmented ions. ESI-MS is in use for many different types of mass spectrometers, both offline and coupled to high-pressure liquid chromatography (HPLC). The advantages for coupling to liquid chromatography (LC), especially at low flow in small bore columns for HPLC, is the increase in sensitivity when allowing instruments to analyze fractions of one sample separated over time, still allowing for the use of small sample volumes. A typical simple HPLC setup for a proteomics mass

spectrometer is a reverse phase non-polar column of silica –C18, with a mobile phase being polar to nonpolar gradients of water and acetonitrile (ACN), added a minute amount of formic acid. The chromatography column ends directly to the ESI, and the chromatographed sample is continuously injected in the mass spectrometer. The typical gradient time is some 60-80 minutes, gaining a separation over time for the content of the sample – thus allowing the MS instrument to thoroughly analyze, fragment and identify peptides as they appear eluted from the chromatography column.

Figure 7. Mass spectrometry principle

There are several different forms of mass spectrometers, with different qualities [99, 106, 107]. In principle, the mass spectrometer consists of three parts, the ion source, the mass analyzer and the detector (see figure X).

ESI and MALDI are ion sources, converting a peptide from a solid or solubilised form to a gaseous charged molecule capable of “flying” in electric fields in vacuum. There are several other modalities than ESI and MALDI in ion sources, but for most mass spectrometers in proteomics, ESI is by far the most predominant.

Mass analyzers separate ions depending on mass per charge (m/z) in an electric field and can perform different functions, depending on architecture. The range is from simple separators of ions based on mass per charge to the more complex combined units where a mass spectrometer is coupled to collision cells selecting and fragmenting separated ions (MS-MS). This fragmentation divides the peptide in smaller random fragments, mainly with the purpose on basis in statistics to reconstitute the sequence of

amino acids in the peptide, as well as to free of reporters for labelled tags for analysis, further discussed under “quantitative proteomics”.

The detector is recording individual m/z from the separated peptides, and also the number ions hitting the detector, thus providing grounds for determining relative intensity of ions present.

Figure 8. Peptide fragmentation patterns. The figure represents fragmentation patterns in principle in an amino acid chain, where b/y – ions, separation in the (C=O)-N-H amide bond is the most common (figure from Wikimedia.com under creative commons licence, adapted from Roepstorff [108]).

The principle of quadropole mass spectrometers [107] is that a filter passes ions through chambers limited of charged rods, where ions can be selected to pass through on basis of ion resonance of m/z. Only specific m/z molecules are able to pass, and can therefore be separately measured by a detector behind the quadropole. A quadropole

cell in this instrumental set-up can also be used as a trap, confining selected ions to a limited space before being passed further within the instrument.

Single quadropoles itself are not very specific, but they are often combined with other modalities or placed in sequence. Linear trap quadropole mass analyzers (LTQ) consist in principle of three quadropoles in a row. LTQ instruments are able to select for specific masses in the first quadropole, colliding and fragmenting in a second quadropole and again analyzing the fragments of masses selected in a third. This gives information of the content of specific peaks identified in the mass spectrum, giving opportunities to follow specific fragments of peptides – a very useful feature for selected reaction monitoring (SRM) [109, 110], further discussed in the section

“quantitative proteomics”. SRM is a label free technique to follow specific fragments;

it is not necessary to “scan” the whole mass spectrum. This gives an increase in instrument sensitivity of one to two orders of magnitude [109].

Figure 9. Triple quadropole experiment linked to LC-ESI in a SRM setup. The first quadropole selects peptides as a filter for further analyses, the second quadropole fragments the selected peptides, and the third select fragments for analysis. Figure adapted from Lange [109] under Creative Commons licence.

Time of Flight (TOF) mass spectrometry is a more specific mass analyzer than quadropoles. Ions enter and the m/z is measured on basis of their flight time through a charged vaakum tube. This is a versatile and stable mass analyzer and is the common mass analyzer for MALDI.

Detectors of TOF and trap-instruments are of electron multiplier types, and they typically add up information with a trap to detect the m/z and relative intensity of a molecule. Such detectors work by principle of an emissive material, such as if a charged particle hits the detector, several electrons might be emitted and each lead to new emissions, causing the generation of a detectable current on the end plate [111].

A Fourier Transform Ion Cyclotron Resonance Mass Spectrometer (FT-ICR-MS) applies another principle of detecting masses. Ions are gathered into an ionization chamber and are exited in a magnetic field cyclotron. Detectors measure the

movement or orbit of ions in the cyclotron. The signal decay from excitations of ions, known as free inductance decay (FID), is detected and transformed to resonance frequencies by Fourier transform and these frequencies are proportional to

mass/charge. These mass analyzers have high resolution and can be used for a precise measurement of masses. Orbitrap MS applies the same principle as FT-ICR-MS, using an oscillating electric field instead of an expensive superconducting magnet [90]. The principle is to inject ions to the Orbitrap observing the stabilizing of an orbit based on electrostatic attraction and centripetal acceleration, depending on m/z. The m/z of trapped ions is thus easily deductible by Fourier transform detected ion orbits, where resolution is increased on every ion “passing” the detector in its orbit. Orbitrap MS instruments are well described by Hu et al [112] and Olsen et al [113].

2.10.1 Challenges for mass spectrometry in proteomics

The difference in concentration between the most abundant proteins and proteins of possible interest that exist in lower concentration is a large challenge in proteomics.

For any instrumental set-up, the difference between the most abundant protein in concentration and the one with lowest concentration detectable is termed the dynamic range. The dynamic range is often measured in orders of magnitude, where most instrument set-ups can handle 10⁴-10⁶, while the biological systems subject for

methodology relevant for this thesis, cerebrospinal fluid and plasma is up to 10¹⁰-10¹¹. This can be exemplified further. Using plasma as an example, the concentration of one single protein – albumin, makes up for about 2/3 of the total protein content of about 60 g/litre. Added up, immunoglobulins, blood coagulation factors fibrinogen and lipoproteins together with albumin make up for 99 % of the protein content in plasma.

Therefore, a strategy is needed to overcome that problem.

Spreading the proteins entering the mass spectrometer over time by chromatography is a strategy much employed to allow the instrument to focus at a few polypeptides at a time. Still, the abundant proteins tend to dominate any sample due to their sheer numbers. Several approaches have been applied in proteomics; unspecific depletion of abundant proteins, specific immunoassay targeting of selected proteins to selectively remove abundant proteins, and extensive fractionation prior, to analysis.

Removal of high abundant proteins from body fluids can be conducted in several different ways, but the most recognised approach is to use columns with antibodies agains the most abundant proteins. Examples of such columns are MARS (multiple removal affinity system) from Agilent, Seppro IgY from Genway Biotech, and Proteoprep from Sigma. Each vendor has several different and more complex columns absorbing more high-abundant proteins from a sample. For the future, this might be expected to further increase the feasibility of analysis of low-abundant proteins.

2.10.2 Alternatives to mass spectrometry in proteomics

Antibodies can also target directly the specific protein in question, by methods such as Enzyme-linked immunosorbent assay (ELISA) and western blot. The principle of ELISA is to create an assay where the substance in question is attached to a surface or solid support by antibodies; a new “detection” antibody is then attached with a reporter, such as an enzyme so that, by measuring enzyme activity, the amount of a protein is reported by indirect measures. This is a useful method for control procedures

and routines, but not in the discovery phases. Western blot is more useful in proteomics, especially in hypothesis-driven proteomics. A western blot consist of a separation on a gel, transfer to a membrane, incubation with primary antibody against the protein in question, and later incubation with secondary antibody against the first, attached to a reporter enzyme. Gel electrophoresis, as previously mentioned, has been the traditional alternative to mass spectrometers, but is on a decline due to sensitivity and reproducibility issues.